Index

Introduction

• Ever since the development of organized policing in America, the U.S. enters modern times in which it must couple with the fact that its structures of law enforcement were built to severely restrict humans rights towards people of color, particularly Black people. As protest numbers towards fighting against racial inequities, police brutality, and civil rights issues increase, it becomes all the more imperative to reassess police agencies and prisons across the country to target the many cracks that remain in the justice system. The Vera Institute has worked towards compiling information on incarceration trends since 1978 through the use of COJ and ASJ datasets in an effort to reveal disproportionalities in jail/prison populations, admission rates, release rates, and more at the county level. Analysis in this report makes use of said data.
• The following report aims to highlight differences between white and black jail/prison populations in particular. Variables of interest include: black jail population, white jail population, total black population ages 15-64, total white population ages 15-64, black jail admissions, white jail admissions, total jail admissions, black prison populations, and white prison populations. Numerous media sources tend to form comparisons between whites and blacks in America regarding incarceration, which resulted in personal curiosity to determine if most disparities were caused by the most populous states, if admission rates were fairly equal between white and black in comparison to their respective population totals by race as well as by total admission rates in general, and if there is reason to suspect of underreporting or a lack of transparency in the data. As a result, incarceration will be measured based off of jail populations and prison populations averaged over the years 2000-2018 from the culmination of all county data for state-level analysis, as well as by admission number totals using the same process. This report hopes to answer some of the following questions or curiosities given the variables of choice:
  • Have there been any dramatic changes in prison population by race in the most populous states?
  • Are the most populous states “to blame” for incarceration disparities, or should efforts be focused elsewhere?
  • Relative to a race’s population in a state, is the admission rate to prison excessively high or low?
  • For each black prison admission in a state, how many white prison admissions are there?
  • Are there certain states that lack accountability in reporting jail data? Why is that?
  • When factoring in the racial makeup by percentage of the U.S., are jail populations on average evenly distributed between whites and blacks?

Summary Information

• By focusing on the variables listed above, 5 summary statistics were computed with an emphasis on the above questions related to the presence of null values, differences in jail population by race over time, and largest proportional admission rates. The first value within the summary list is “largest_avg_black_adm”, which refers to the largest value for average black prison admissions divided by the average total black population from ages 15 to 64 to be proportional to a state’s general black population. This value was determined to be 0.04089634 and corresponds to the next item in the list “state_with_largest_avg_black_adm”, referring to the state with the greatest average proportional black prison admissions, which happens to be Arkansas. This became obvious during the process of creating plots, as this outlier greatly distorted the line of best fit and was considerably distant from the rest of the U.S., indicating that there may be severe issues present in incarceration rates for people of color in AK that may be worth investigating. The second value is named “num_states_null_white_prison_pop”, which details the exact number of states that fail to report white prison populations between 2000 to 2018, as this was indicated by containing only null values when running a for loop, and was assigned to the number 0 as only states with completely null data returned sums of 0. This value is 11L, indicating that 11 states have failed to report any indication of prison population makeup by race for whites in a timespan of 18 WHOLE years. This value corresponds to the third summary statistic “num_states_null_black_prison_pop”, which also is 11L as the same states have failed to record any values for the number of black prisoners in the last 18 years in ANY county for that matter. This means that approximately 22% of our dataset (51 states) may be lacking greatly in prison accountability and transparency behind their methods of law enforcement or imprisonment, whether that is being intentionally withheld or as a result of ignorance. Either way, it is all the more important to begin to emphasize records efforts in such states. The fourth and fifth summary statistics also correspond in variables of interest, given by the names “largest_white_jail_pop_diff” and “largest_black_jail_pop_diff”. These values aim to determine the most dramatic changes in a broad view between the jail population in 2000 versus the jail population in 2018 and see if these values are on par between whites and blacks. The largest difference in white jail populations was 9560.2, while the largest difference in black jail populations was 1571.64. The following differences also correspond to state names (listed in the summary list), which were Tennessee and North Carolina. The rise in white individuals in jail is particularly dramatic, which may prompt further research to determine if this is a result of population increases, demographic changes, specific policies made to policing, or excessive incarceration rates in general. As for the black population increase, the lower value may be difficult to comprehend given an emphasis on racial disparities, but this aligns with the time series plot which has indicated a slight downwards trend in black jail populations for the top 6 states, which could also be a key point to analyze and correspond to certain policy changes or demographic shifts.

Plots

Trends Over Time

• Description: The above plot displays the average total black jail population summed between all counties within the 6 states with the largest population in the U.S. between the years of 2000 to 2018. The purpose of this chart was to identify if there have been sudden rises or declines in black jail population within the 2000s and if the largest states have the greatest impact on racial disparities in incarceration rates in comparison to the white jail population displayed below. What is most intriguing about this graph and resulted in its presence in the report is the consistent downwards trend (for the most part), particularly after the year of 2008 or so. Florida dropped by approximately 7,000 individuals between the years of 2008 to 2013. Another key detail to note is that the number of black individuals in jail in New York and Illinois specifically exceed that of white individuals in jail, even after the downwards trend. Beyond this, there is some alignment in the minor peaks seen throughout the years for multiple states between rises in black populations and white populations, but the disparity is prevalent when taking into consideration the proportion of black people in each of these states.

• Description: The above plot displays the average total white jail population summed between all counties within the 6 states with the largest population in the U.S. between the years of 2000 to 2018. The purpose of this chart was to identify if there have been sudden rises or declines in white jail population within the 2000s and if the largest states have the greatest impact on racial disparities in incarceration rates in comparison to the black jail population displayed above. What was most intriguing about this graph and resulted in its presence in the report is the consistent stability in jail populations for New York and Illinois, both at levels considerably lower than black jail populations and relatively small in proportion to the overall state populations. Otherwise, there is a lack of consistency in the rise and fall of white jail populations, and instead less variation than that seen in black jail populations. What is consistent is that, just as the graph for black populations, the ordering of states with most to least individuals over time is mostly the same. It should also be noted that the maximum y values for this plot exceed the previous one, but it would be fairer to acknowledge this as a result of the large proportion of white individuals in these states as well as in the U.S. as a whole.

Variable Comparisons

• Description: The above plot displays the average number of white prison admissions proportional to a state’s average white population versus the average number of black prison admissions proportional to a state’s average black population. These values were determined by the sums of all county data of a state over the years 2000 to 2018. A line of best fit has also been included using linear regression. The data seems to have a fairly strong correlation, and the plot’s purpose was to determine if prison admission rates were relatively equal or changed in similar ways after adjusting for total population as a fair metric of seeing what percentage of the total population for a certain race is really being admitted into jail. With a slope of 3.46, there seems to be a strong disparity towards the admissions of black individuals compared to white individuals per state, with the x-axis only ranging from 0 to around 0.004 while the y-axis ranges from 0 to slightly above 0.0015. The states with admission rates for blacks significantly greater than whites were mostly beyond the top 6 states with the highest populations and instead correspond to many states that typically have a majority vote in favor of the Republican Party, which may have some influence over policies that exist in the policing systems of those states.

• Description: The above plot displays the average number of black prison admissions proportional to a state’s average total prison admissions versus the average number of white prison admissions proportional to a state’s average total prison admissions. These values were determined by the sums of all county data of a state over the years 2000 to 2018. A line of best fit has also been included using linear regression. The data seems to have a weaker correlation and is fairly scattered across the U.S., and this plot had a similar purpose as the last to identify any trends present as a result of trying to “equalize” black prison admissions and white prison admissions by making them as proportions relative to total prison admissions. However, it is apparent that this is not as well of a metric as the previous was based on population within each state, which may actually be a result of the fact that the total prison admissions themselves are skewed towards particular racial groups. There may be a slight negative trend present throughout the U.S., where increases in black prison admissions correspond with decreases in white prison admissions, but this seems more unlikely and is an indication that a different approach may be needed for this problem, or a different fit rather than the least squares method. This plot serves as evidence that not all proportions are useful or lead to equalized results, and that our preexisting beliefs related to research topics do not always have to end up being proven.

Maps

• Description: The above regional plot displays the average number of black prisoners across various counties within the 50 states between the years of 2000 to 2018. An important note to this graph is that null values, or an absence of data for the number of black prisoners, is indicated by a value of 0. There are a total of 11 states that fall into this category. The purpose of the following visualization is to determine if black prison populations are simply correlated to states with high populations in general, or if there are apparent disparities regionally when contrasted with white prison populations. One key observation to note is that the largest number of prisoners for this graph reaches 40,000 despite the fact that the black population in the U.S. overall is considerably lower than that for the white population. The below graph corresponds to the above graph in that three of the states with the largest populations also have the largest black prison populations and white prison populations, which can be a combination of large population size as well as skewed excess of incarcerations of black individuals. To observe from a case-study basis, Alabama has an average of 14000 black prisoners compared to 9210 white prisoners from the below graph. It appears that in both graphs data reporting is consistent in most places excluding the upper northwest portions of the U.S.. Unfortunately, it does appear that the area known as the “black belt” where there is a higher proportion of black individuals in the U.S. exceeds the white population. Simply when given the perspective that only around 13.4% of the U.S. is black compared to 61.6% white, there needs to be a tremendous overhaul in policies related to incarcerations and systemic efforts to eradicate the pervasive racial bias that remains.

• Description: The above regional plot displays the average number of white prisoners across various counties within the 50 states between the years of 2000 to 2018. An important note to this graph is that null values, or an absence of data for the number of white prisoners, is indicated by a value of 0. There are a total of 11 states that fall into this category. There are a total of 11 states that fall into this category. The purpose of the following visualization is to determine if white prison populations are simply correlated to states with high populations in general, or if there are apparent disparities regionally when contrasted with black prison populations. One key observation to note is that the largest number of prisoners for this graph reaches around 35,000 white prisoners despite the fact that the black population in the U.S. overall is considerably lower than that for the white population. To observe from another case-study basis, California, Texas, and Florida all have average white prisoner populations that are less than their black prisoner populations, differing in population by 5-8k more black individuals. Towards some of the west and midwest, multiple states exceed in white prisoner populations, which may be more as a result of incredibly low black populations, or improved policies in incarceration methods, and may require further investigation. An overall takeaway here, however, is that a large proportion of states contain prisoner populations with more blacks than whites, despite the differences in these racial proportions within the nation, making it all the more imperative to utilize these sources as a means to push for public policy.